Medulloblastoma, a tumor of the central nervous system, is a common and frequently lethal tumor of childhood. Abnormalities in Sonic Hedgehog (SHH) and Wingless signaling contribute to genetics, however these pathways are represented in a minority of tumors. The proto-oncogene Mycn is the major factor driving proliferation in the developing cerebellum, and is expressed at high levels in SHH-associated medulloblastoma. A role for Mycn in the pathogenesis of medulloblastoma in the absence of SHH aberrations is supported by three independent studies showing a majority of human medulloblastomas express Mycn, whereas Mycn is not expressed in normal cerebellum after fetal stages. We hypothesize that aberrant expression of Mycn contributes to the pathogenesis of medulloblastoma, and that targeted therapies against Mycn will show efficacy in this disease. Our long term objectives are to clarify the role of Mycn in the initiation and progression of medulloblastoma, and to determine the impact of therapies directed against Mycn in murine and human medulloblastoma. We will characterize a genetically engineered mouse (GEM) model for medulloblastoma developed in our laboratory that co-expresses both Mycn and luciferase in the CNS, and that differs from existing GEM models for medulloblastoma in that p53 mutation (which is uncommon in human tumors) is not required to achieve high penetrance. Mycn is stabilized through activation of the lipid kinase PI3K. We have shown that small molecule inhibitors of PI3K lead to degradation of this oncoprotein in-vitro and in-vivo. We will therefore test the role of Mycn in maintaining tumors and in supporting tumor vasculature by treating MYCN-driven medulloblastomas using doxycyline to turn off MYCN transcriptionally, or using small molecule PI3K) inhibitors to degrade Mycn. We will compare genetic and epigenetic abnormalities between murine and human tumors, including microRNA profiling, and we will treat cultured and xenotransplated tumorspheres derived from primary human medulloblastomas to assess the importance of Mycn as a therapeutic target in human medulloblastoma. Aim 1. Evaluation of mice transgenic for Glt1-tTA:TRE-MYCN/Luc as a model for human medulloblastoma. Aim 2. Preclinical testing of isoform-selective PI3K inhibitors in mice transgenic for Glt1-tTA:TREMYCN/Luc, and in human medulloblastoma tumorspheres. Aim 3. Use of PI3K inhibitors to address a role for Mycn and microRNA targets regulated by Mycn, in maintaining medulloblastoma angiogenesis in vivo.